© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 31693
    Prev
    Next
    INDEX
    Explanations

    instances of "interview" and related contexts, indicating discussions or conversations reported in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    564
    +0.12
    0.6%
    306
    +0.11
    0.5%
    224
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    224
    +0.12
    0.05
    729
    +0.11
    0.03
    317
    +0.10
    0.03
    Negative Logits
     CoC
    -0.75
     semblance
    -0.67
     Documentation
    -0.65
     resemblance
    -0.60
    HK
    -0.60
    ulum
    -0.58
     EntityItem
    -0.57
     Gum
    -0.56
     Gry
    -0.56
     Um
    -0.55
    POSITIVE LOGITS
    interrupted
    0.98
     airing
    0.95
     aired
    0.93
     broadcast
    0.92
     published
    0.83
     TODAY
    0.79
     afterward
    0.78
     streamed
    0.77
     conducted
    0.76
    éĸ
    0.74
    Activations Density 0.269%

    No Known Activations